Exploring the structure of physical systems, from microscopic scales to the cosmos.
I'm a Physics PhD researcher interested in how mathematics and computation help us uncover structure in complex systems. My work brings together topological data analysis (TDA), machine learning, numerical simulation, and physical interpretation.
A question that motivates much of what I do:
What can the shape of data reveal about the physics that generated it?
I care about more than predictive accuracy: I enjoy developing methods that are interpretable, physically meaningful, and reproducible.
| ⚛️ Microscopic | 🌊 Mesoscopic | 🌌 Astronomical |
|---|---|---|
| Spectral structure, condensed matter, and materials | Nonlinear dynamics, fluid and magnetohydrodynamic systems | Stellar oscillations, gravitational waves, and galaxy morphology |
| Topology of spectra & physical observables | Patterns, instabilities & dynamics | Structure & inference from astronomical data |
Methods connecting these scales: persistent homology · scientific machine learning · time-series analysis · numerical modeling · statistical inference.
A selection of my open-source work, spanning research pipelines and computational physics:
| Topology & data-driven physics | Numerical modeling & simulations |
|---|---|
| Gravitational waves at low SNR Exploring persistent-homology features for signal detection in noisy time series. |
PyMoS² — Static stellar models Numerical modeling of stellar structure, with an emphasis on transparent, educational code. |
| Galaxy morphology with TDA + SSL Topological and contrastive representations for analyzing galaxy morphologies. |
Brio–Wu shock tube: TDA + CNN Using topological descriptors to study neural-prediction behavior in MHD simulations. |
Other tools & teaching repositories
- PET.py — Jupyter notebooks on computational physics, astronomy, and applied mathematics.
- Intro_ML — Hands-on material for an introductory machine-learning minicourse.
- AI-Gen-RAG Bibliographic Review Assistant — Exploring retrieval-augmented workflows for literature review.
Explore the rest of my work in all repositories.
- A machine learning and neural network approach to flood forecasting in the Wupper river · Anais da Academia Brasileira de Ciências (2026).
- Computational Modeling of a Star: A Theoretical and Numerical Approach · Revista Brasileira de Ensino de Física (2025).
- Generalization to d-dimensions of a fermionic path integral for exact enumeration of polygons on hypercubic lattices · Scientific Reports (2024).
- The limb darkening via transit modeling and Bayesian inference · Revista Brasileira de Ensino de Física (2022).
For a fuller publication record, visit my ORCID or Lattes CV.
| Scientific computing | Machine learning & data | Research workflow |
|---|---|---|
Python · NumPy · SciPy · pandas |
scikit-learn · TensorFlow/Keras |
Git · Jupyter · LaTeX |
| Numerical methods · ODEs/PDEs · signal processing | TDA · time-series analysis · scientific visualization | Matplotlib · Plotly · Streamlit |
My usual priorities: clear scientific assumptions, well-documented experiments, and results that others can reproduce.
I enjoy making quantitative ideas approachable without giving up mathematical rigor. My teaching material includes computational-physics notebooks, an introductory machine-learning minicourse, and Python workshops for students in science and engineering.
Whenever possible, I share the code and learning resources openly so that other people can adapt them.
Curiosity is a good starting point. Reproducibility makes it science.
Thanks for visiting! Explore a project, open a discussion, or connect through the links above. 🌺